Matthew Rabin has a devastating attack on expected utility theory in
Econometrica (V.65 #5). Although the idea has sort of been known for
some time Rabin brings it home that any expected utility functin which
can explain risk aversion for small gambles offers utterly implausible
predictions for large gambles and any function which can explain
plausible risk aversion for large gambles predicts risk neutrality for
relatively small gambles - which is not what we observe.
For example, Rabin shows that if an expected utility maximizer turns
down the 50,50 bet of
lose $10/gain $11 then he must always turn down the 50, 50 bet of lose
$100 gain / *any amount*!
or if (same as above)
lose $1000/ gain $1250 then he must always turn down the 50,50 bet of
lose $6000 / gain *any amount*.
What is especially interesting is that the proof of these
assertions is not terribly difficult and has been sitting in front of
everyone for years.
You can find Rabin's papers on this on his home page at
http://elsa.berkeley.edu/users/rabin/
(got to publications in PDF format to find)
"Risk Aversion, Diminishing Marginal Utility, and Expected-Utility
Theory: A Calibration Theorem," May 1999
Abstract (HTML) or Full paper (PDF)
and
"Diminishing Marginal Utility of Wealth Cannot Explain Risk Aversion,"
August 1999
Alex
--
Dr. Alexander Tabarrok
Vice President and Director of Research
The Independent Institute
100 Swan Way
Oakland, CA, 94621-1428
Tel. 510-632-1366, FAX: 510-568-6040
Email: [EMAIL PROTECTED]
